Re: [Qemu-devel] [PATCH COLO-Frame v12 28/38] COLO: Process shutdown command for VM in COLO state

[Hailiang Zhang]

On 2015/12/15 19:31, Dr. David Alan Gilbert wrote:
> * zhanghailiang (address@hidden) wrote:
> > If VM is in COLO FT state, we should do some extra work before normal shutdown
> > process. SVM will ignore the shutdown command if this command is issued directly
> > to it, PVM will send the shutdown command to SVM if it gets this command.
> >
> > Cc: Paolo Bonzini <address@hidden>
> > Signed-off-by: zhanghailiang <address@hidden>
> > Signed-off-by: Li Zhijian <address@hidden>
> > ---
> >   include/sysemu/sysemu.h |  3 +++
> >   migration/colo.c        | 25 +++++++++++++++++++++++--
> >   qapi-schema.json        |  4 +++-
> >   vl.c                    | 26 ++++++++++++++++++++++++--
> >   4 files changed, 53 insertions(+), 5 deletions(-)
> >
> > diff --git a/include/sysemu/sysemu.h b/include/sysemu/sysemu.h
> > index 3bb8897..91eeda3 100644
> > --- a/include/sysemu/sysemu.h
> > +++ b/include/sysemu/sysemu.h
> > @@ -52,6 +52,8 @@ typedef enum WakeupReason {
> >       QEMU_WAKEUP_REASON_OTHER,
> >   } WakeupReason;
> >
> > +extern int colo_shutdown_requested;
> > +
> >   void qemu_system_reset_request(void);
> >   void qemu_system_suspend_request(void);
> >   void qemu_register_suspend_notifier(Notifier *notifier);
> > @@ -59,6 +61,7 @@ void qemu_system_wakeup_request(WakeupReason reason);
> >   void qemu_system_wakeup_enable(WakeupReason reason, bool enabled);
> >   void qemu_register_wakeup_notifier(Notifier *notifier);
> >   void qemu_system_shutdown_request(void);
> > +void qemu_system_shutdown_request_core(void);
> >   void qemu_system_powerdown_request(void);
> >   void qemu_register_powerdown_notifier(Notifier *notifier);
> >   void qemu_system_debug_request(void);
> > diff --git a/migration/colo.c b/migration/colo.c
> > index f4bb661..a094991 100644
> > --- a/migration/colo.c
> > +++ b/migration/colo.c
> > @@ -231,6 +231,7 @@ static int colo_do_checkpoint_transaction(MigrationState *s,
> >                                             QEMUSizedBuffer *buffer)
> >   {
> >       int ret;
> > +    int colo_shutdown;
> >       size_t size;
> >       QEMUFile *trans = NULL;
> >
> > @@ -258,6 +259,7 @@ static int colo_do_checkpoint_transaction(MigrationState *s,
> >           ret = -1;
> >           goto out;
> >       }
> > +    colo_shutdown = colo_shutdown_requested;
> >       vm_stop_force_state(RUN_STATE_COLO);
> >       qemu_mutex_unlock_iothread();
> >       trace_colo_vm_state_change("run", "stop");
> > @@ -311,6 +313,15 @@ static int colo_do_checkpoint_transaction(MigrationState 
> > *s,
> >           goto out;
> >       }
> >
> > +    if (colo_shutdown) {
> > +        colo_put_cmd(s->to_dst_file, COLO_COMMAND_GUEST_SHUTDOWN);
> > +        qemu_fflush(s->to_dst_file);
> > +        colo_shutdown_requested = 0;
> > +        qemu_system_shutdown_request_core();
> > +        /* Fix me: Just let the colo thread exit ? */
> > +        qemu_thread_exit(0);
> > +    }
> > +
> >       ret = 0;
> >       /* Resume primary guest */
> >       qemu_mutex_lock_iothread();
> > @@ -370,8 +381,9 @@ static void colo_process_checkpoint(MigrationState *s)
> >           }
> >
> >           current_time = qemu_clock_get_ms(QEMU_CLOCK_HOST);
> > -        if (current_time - checkpoint_time <
> > -            s->parameters[MIGRATION_PARAMETER_X_CHECKPOINT_DELAY]) {
> > +        if ((current_time - checkpoint_time <
> > +            s->parameters[MIGRATION_PARAMETER_X_CHECKPOINT_DELAY]) &&
> > +            !colo_shutdown_requested) {
> >               int64_t delay_ms;
> >
> >               delay_ms = s->parameters[MIGRATION_PARAMETER_X_CHECKPOINT_DELAY] -
> > @@ -442,6 +454,15 @@ static int colo_wait_handle_cmd(QEMUFile *f, int 
> > *checkpoint_request)
> >       case COLO_COMMAND_CHECKPOINT_REQUEST:
> >           *checkpoint_request = 1;
> >           return 0;
> > +    case COLO_COMMAND_GUEST_SHUTDOWN:
> > +        qemu_mutex_lock_iothread();
> > +        vm_stop_force_state(RUN_STATE_COLO);
> > +        qemu_system_shutdown_request_core();
> > +        qemu_mutex_unlock_iothread();
> > +        /* the main thread will exit and termiante the whole
>
> Typo 'termiante'

> > +        * process, do we need some cleanup?
> > +        */
> > +        qemu_thread_exit(0);
>
> Yes, I'm not sure how much real cleanup you need during shutdown;
> I wonder how a shutdown will look to the management layers above;
> if they don't realise it's a shutdown they might try and do a failover
> when one side exits.

I have tested this case for several times, it seems worked well.

> >       default:
> >           return -EINVAL;
> >       }
> > diff --git a/qapi-schema.json b/qapi-schema.json
> > index f6ecb88..b5b1a02 100644
> > --- a/qapi-schema.json
> > +++ b/qapi-schema.json
> > @@ -754,12 +754,14 @@
> >   #
> >   # @vmstate-loaded: VM's state has been loaded by SVM.
> >   #
> > +# @guest-shutdown: shutdown require from PVM to SVM
> > +#
> >   # Since: 2.6
> >   ##
> >   { 'enum': 'COLOCommand',
> >     'data': [ 'checkpoint-ready', 'checkpoint-request', 'checkpoint-reply',
> >               'vmstate-send', 'vmstate-size','vmstate-received',
> > -            'vmstate-loaded' ] }
> > +            'vmstate-loaded', 'guest-shutdown' ] }
> >
> >   ##
> >   # @COLOMode
> > diff --git a/vl.c b/vl.c
> > index fca630b..1a61300 100644
> > --- a/vl.c
> > +++ b/vl.c
> > @@ -1636,6 +1636,8 @@ static NotifierList wakeup_notifiers =
> >       NOTIFIER_LIST_INITIALIZER(wakeup_notifiers);
> >   static uint32_t wakeup_reason_mask = ~(1 << QEMU_WAKEUP_REASON_NONE);
> >
> > +int colo_shutdown_requested;
> > +
> >   int qemu_shutdown_requested_get(void)
> >   {
> >       return shutdown_requested;
> > @@ -1767,6 +1769,10 @@ void qemu_system_guest_panicked(void)
> >   void qemu_system_reset_request(void)
> >   {
> >       if (no_reboot) {
> > +        qemu_system_shutdown_request();
> > +        if (!shutdown_requested) {/* colo handle it ? */
> > +            return;
> > +        }
> >           shutdown_requested = 1;
>
> Do we still need that 'shutdown_requested = 1'  - it's already
> true at this point or it returned?

No, it is useless, i will remove it.

> >       } else {
> >           reset_requested = 1;
> > @@ -1840,14 +1846,30 @@ void qemu_system_killed(int signal, pid_t pid)
> >       qemu_notify_event();
> >   }
> >
> > -void qemu_system_shutdown_request(void)
> > +void qemu_system_shutdown_request_core(void)
> >   {
> > -    trace_qemu_system_shutdown_request();
> >       replay_shutdown_request();
> >       shutdown_requested = 1;
> >       qemu_notify_event();
> >   }
> >
> > +void qemu_system_shutdown_request(void)
> > +{
> > +    trace_qemu_system_shutdown_request();
> > +    /*
> > +    * if in colo mode, we need do some significant work before respond to the
> > +    * shutdown request.
> > +    */
> > +    if (migration_incoming_in_colo_state()) {
> > +        return ; /* primary's responsibility */
> > +    }
> > +    if (migration_in_colo_state()) {
> > +        colo_shutdown_requested = 1;
> > +        return;
> > +    }
>
> Try to move most of this into migration/colo*.c ;
> here you could just do:
>      if (colo_shutdown()) {
>          return;
>      }
>
> it's best to keep vl.c as simple as possible.

Yes, it is reasonable. I will fix it in next version.

> Dave
>
> > +    qemu_system_shutdown_request_core();
> > +}
> > +
> >   static void qemu_system_powerdown(void)
> >   {
> >       qapi_event_send_powerdown(&error_abort);
> > --
> > 1.8.3.1
> --
> Dr. David Alan Gilbert / address@hidden / Manchester, UK
>
> .